effect of therapeutic heparin vs prophylactic heparin on

1

Effect of Therapeutic Heparin vs Prophylactic Heparin on Death, Mechanical Ventilation

or Intensive Care Unit Admission in Moderately Ill Ward Patients with COVID-19: The

RAPID Randomized Clinical Trial

Sholzberg et al.

2

Table of Contents

1. Investigators and Committee Members ....................................................................................4

2. Funding Agencies....................................................................................................................7

3. Supplementary Methods ..........................................................................................................8

3.1 Eligibility Criteria.......................................................................................................................... 8

3.2 Description of Therapeutic Heparin vs. Prophylactic Heparin ........................................................ 8

3.3 Primary Outcome .......................................................................................................................... 9

3.4 Pre-specified Outcome Definition for Organ-support Free Days ................................................... 10

3.5 Sample Size Considerations ......................................................................................................... 10

3.6 Extended Description of Statistical Methods ................................................................................ 11

3.7 Adjudication ................................................................................................................................ 11

3.8 Data and Safety Monitoring Board ............................................................................................... 14

3.9 Trial Administration .................................................................................................................... 15

4. Supplementary Figures and Tables ........................................................................................ 16

Figure S1. Trial Schematic ................................................................................................................ 16

Table S1. Recruitment Numbers By Site ............................................................................................ 17

Table S2. Means and Medians and Measures of Dispersion of Continuous Baseline Characteristics ... 18

Table S3. Duration and Type of Heparin Used ................................................................................... 19

Table S4. Concomitant Treatments Received ..................................................................................... 20

Table S5. Means and Medians and Measures of Dispersion of Continuous Outcomes ........................ 21

Table S6. Primary Cause of Death ..................................................................................................... 22

Figure S2. Time-to-event Analyses of the Primary Composite Outcome ............................................ 23

Figure S3. Time-to-event Analyses of All-cause Death ...................................................................... 24

Figure S4. Time-to-event Analyses of Mechanical Ventilation ........................................................... 25

Figure S5. Time-to-event Analyses of Intensive Care Unit Admission ............................................... 26

Table S7. Thromboembolism............................................................................................................. 27

Table S8. ISTH Major Bleeding Events ............................................................................................. 28

Table S9. Bleeding Events by Concomitant Treatments Received ...................................................... 29

Figure S6. Post-Hoc Subgroup Analysis of All-Cause Death.............................................................. 30

Table S10. Per Protocol Analysis of the Primary Outcome and Its Components ................................. 31

Table S11. Sensitivity Analysis 1 of the Primary Outcome and Its Components ................................. 32



3

Table S14. Intention-to-Treat Analysis of the Primary Outcome and Its Components Adjusted for Age ......................................................................................................................................................... 35

Table S15. Intention-to-Treat Analysis of the Primary Outcome with Time-by-Treatment Interaction 36

Table S16. Intention-to-Treat Analysis of the Primary Outcome Estimating Risk Differences ............ 37

Table S17. Sensitivity Analyses of D-Dimer Levels at Day 2 ............................................................. 38

Table S18. Study Drug Not Received as Allocated within the First 48 hours ...................................... 39

Table S19. Study Drug Change within the First 48 hours With and Without Clear Clinical Indication 40

Table S20. Primary and Secondary Outcomes until Hospital Discharge ............................................. 41

5. References............................................................................................................................. 42

4

1. Investigators and Committee Members

Co-Principal investigators: Michelle Sholzberg, Peter Jüni, Mary Cushman

Steering Committee: Michelle Sholzberg (Chair), Peter Jüni, Mary Cushman, Musaad AlHamzah, Lisa Baumann

Kreuziger, Andrew Beckett, Marc Carrier, Bruno R. da Costa, Michael Fralick, Paula D. James, Agnes Y.Y. Lee,

David Lillicrap, Saskia Middeldorp, Fionnuala Ní Áinle, Elnara Márcia Negri, Grace H. Tang, Kevin E. Thorpe

Data and Safety Monitoring Board: Andreas Laupacis (Chair), Yulia Lin, Andrew Day, Bram Rochwerg

Medical Monitors: Eric K. Tseng, Gloria Lim

Independent Event-Adjudication Committee: Jameel Abdulrehman (English), Mansour Gergi (English), Rodrigo

Hidd Kondo (Portuguese), Bruna Mamprim Piloto (Portuguese)

Study Investigators and Coordinators:

Brazil:

Hospital das Clinicas HCFMUSP, Faculdade de Medicina, Universidade de Sao Paulo, SP, BR: Elnara Márcia

Negri, Hassan Rahhal, Carlos Eduardo Pompílio, Guilherme de Abreu Pereira, Fernando Salvetti Valente, Ariel

Fernando Villarroel Agreda, Pablo Andres Munoz Torres, Maíra Oliveira Moraes, Claudia de Lucena Moreira,

Fernando Galassi Stocco Neto, Fabíola Vieira Duarte Baptista, Joanne Alves Moreira, Augusto Séttemo Ferreira,

Paula Frudit, Gabriel Martinez, Heraldo Possolo Souza, Rodrigo Antônio Brandão Neto, Elbio Antonio d'Amico, Eduardo Messias Hirano Padrão, Fernando Sarin da Mota e Albuquerque, Giovanna Villela Zangrossi, Alberto

Kendy Kanasiro, Alissom Vitti Cincoto, Hugo de Souza Reis, Vitor Miyashiro Arias da Silva, Arthur Petrillo

Bellintani, Vivian dos Santos Pereira, Yohan Washington de Oliveira, Bárbara Justo Carvalho, Mariana de Souza

Novaes, Henrique Brito Silveira, Gabriel César Alves de Avelar, Lara Bonanni Mota, Karina Caciola, Matheus de

Arêa Leão Freire Marim, Tales Cabral Monsalvarga, Alexandre Salgado Blanco Santos, Ahmed Haydar, Gabriella

Seixas Sampaio Saraiva, Sabrina Correa da Costa Ribeiro, Julio Flavio Meirelles Marchini, Julio Cesar Garcia de

Alencar, Ana Catharina de Seixas Santos

Canada:

Hôpital Charles Lemoyne, Longueuil, QC: Catherine Sperlich, Caroline Millette, Mathieu Lebeau, Céline Devaux,

Mélina Boutin, Trung Nghia Nguyen, Line Srour, Flavia de Angelis, Mariane Fugulin

William Osler Health System (Brampton Civic Hospital and Etobicoke General Hospital), Brampton and Etobicoke,

ON: Sabrena Tangri, Alexandra Binnie, Shayna A.D. Bejaimal, Andrew Binding, Rosa M. Marticorena, Galo

Ginocchio

Mount Sinai Hospital, Toronto, ON: Michael Fralick, Eric Kaplovitch, Klaudia Rymaszewski, Afsaneh Raissi,

Marcelo Falappa

Trillium Health Partners (Credit Valley Hospital and Mississauga Hospital), Mississauga, ON: Terence Tang, Blair

Ernst, Amna Ali, Martin Romano, Mobina Khurram

St. Joseph's Health Centre, Toronto, ON: Peter Jaksa, Christie Kim, Ajay Kapur, Michelle Edwards, Vidushi Swarup, Bruna Camilotti, Jiten Jani, Jeff Carter

Alberta Health Services (Foothills Medical Centre, Rockyview General Hospital, Peter Lougheed Centre), Calgary,

AB: Deepa Suryanarayan, Mark R. Gillrie, Davinder Sidhu, Traci Robinson

The Ottawa Hospital (General and Civic Campus), Ottawa, ON: Lana Castellucci, Krystina Stutely, Irene Watpool,

Rebecca Porteous

Hôpital du Sacré-Coeur de Montréal, Montréal, QC: Karine Doyon, Kevin Jao, Jean-Samuel Boudreault-Pedneault

Michael Garron Hospital, Toronto, ON: E. Roseann Andreou, Christopher Kandel, Maureen T. Taylor, Wei En

Enoch Choo

5

St. Michael's Hospital, Toronto, ON: Vera Dounaevskaia, Eric K. Tseng, Gloria Lim, Vidushi Swarup, Laura

Parsons, Ann Dowbenka, Gitana Ramonas

University of Alberta Hospital, Edmonton, AB: Cynthia Wu, Sergey Nikitin, Jeffery Patterson, Mark Hnatiuk, Anca Tapardel, Sarah Takach- Lapner, Thirza Carpenter, Lori Rackel, Rebecca Cairns, Jessica Pinder

Southlake Regional Health Centre, Newmarket, ON: Jai Jayakar, Peter Anglin, Catherine McPherson, Liselle

Chiverton, David Barbosa, Shany Loukiantchenko, Yana Shamiss

Hospital Maisonneuve-Rosemont, Montréal, QC: Marie-Pier Arsenault, Danaë Tassy

Nova Scotia Health Authority (Queen Elizabeth II Health Sciences Centre), Halifax, NS: Sudeep Shivakumar,

Mary-Margaret Keating, Sue Pleasance

Ireland:

Mater Misericordiae University Hospital, Dublin and University College Dublin Clinical Research Centre: Fionnuala Ní Áinle, Barry Kevane, Sarah Cullivan, Nick Power, Peter Doran, Kelly Leamy, Aoife Kelly, Conor

Moran, Mairead O'Connor, Aoife McDonnell, Roseanne Boyce, Faiza Sefroun, Rabia Hussain, Patrick Murray,

Anna Malara, Brenda Molloy, Meadbh O’Halloran, Emer Cunningham, Jack Lambert, Aoife Cotter, Brian Marsh,

Gerard Sheehan, Eavan Muldoon, James Woo, Sean Gaine, Deirdre Morley

Saudi Arabia:

King Saud University Medical City, Riyadh: Musaad AlHamzah, Khalid Alayed, Farjah H AlGahtani, Ibrahim

Almaghlouth, Sondus Ata, Fai Alkhathlan, Najma Khalil, Israa Mohamed Hussein, Mohammed Bashir, Ahmed S.

BaHammam, Abdulrahman Alsultan, Hadeel Alkofide, Tariq M Alhawassi

King Fahad Medical City, Riyadh: Mohammed Alsheef, Fahad AlSumait, Abdulhadi M. Alqahtani, Emad K. Zayed, Ammar AlSughayir, Yacoub Abuzied

King Faisal Specialist Hospital, Riyadh: Faris Alomran, Hazzaa AlZahrani, Jawaher Al-Otaibi, Haya Alothaimeen,

Noura Alzannan

United States of America:

Versiti, Milwaukee, Wisconsin: Lisa Baumann Kreuziger, Amer Al Homssi, Haisam Abid, Stephanie Jones,

Shannon Broaddrick, Neha Jain

University of Vermont Medical Center, Burlington, Vermont: Christos Colovos, Mary Cushman, Roz King, Mohit

Jindal

Barnes Jewish Hospital, St. Louis, Missouri: Kristen Sanfilippo, Patty Nieters

United Arab Emirates:

Al Ain Hospital, Al Ain: Mozah Almarshoodi, Muhammad Hammad, Aleeswa Francis Benny, Tariq A. Hamdan,

Suhaib Kamal Ahmed Elobaid, Ibrahim Khafagi, Saima Saeed Ahmed, AbduleRehman AlEssaie, Shamma AlAlawi,

Khloud Bashir, Aysha Abdulla Salem Al Suwaidi, Hiba Ibrahim Khogali Ahmed, Mohamed Milad Ismail, El

Mutasim Ahmed El Faki

Data Management and Coordination Center (DMCC):

Applied Health Research Centre, St. Michael’s Hospital/University of Toronto: Mercy Charles, Alice Dang,

Gurpreet Lakhanpal, Dominic Lee, Prachi Ray, Maria Naydenova, Kosma Wysocki

Hematology-Oncology Research Group, St. Michael’s Hospital/University of Toronto: Michelle Sholzberg, Aziz

Jiwajee, Vidushi Swarup, Bruna Camilotti, Grace H. Tang

Regional Coordination Centers:

US Coordination Center (Versiti, Milwaukee, Wisconsin): Lisa Baumann Kreuziger, Stephanie Jones, Shannon

Broaddrick, Greg Wendling

6

Saudi Arabia Coordination Center (King Saud University Medical City, Riyadh): Musaad AlHamzah, Ahmed S.

BaHammam, Abdulrahman Alsultan, Hadeel Alkofide, Tariq M Alhawassi

European Legal Representative (Mater Misericordiae University Hospital, Dublin): Fionnuala Ní Áinle, Brenda Molloy, Anna Malara, Emer Cunningham, Meadbh O'Halloran, Aoife Cotter, Brian Marsh, Jack Lambert, Gerard

Sheehan, Eavan Muldoon, James Woo, Sean Gaine

Brazil Coordination Center (Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo, São

Paulo): Hassan Rahhal, Elnara Márcia Negri, Carlos Eduardo Pompilio, Fernando S. Valente and Guilherme de

Abreu Pereira.

Statistical Analysis (AHRC): Bruno R da Costa, Kevin E. Thorpe, Fei Zuo, Peter Jüni

Biorepository Coordination Center:

Laboratory for Clinical Biochemistry Research, University of Vermont Larner College of Medicine, Burlington,

Vermont: Mary Cushman, Rebekah Boyle, Elaine Cornell, Debora Kamin Mukaz

7

2. Funding Agencies

The RAPID Trial was funded by:

- Task 54, Defence Research Development Canada, Department of National Defence, Ottawa, Canada

- St. Michael’s Hospital Foundation, Toronto, Canada

- St. Joseph’s Health Centre Foundation, Toronto, Canada

- 2020 TD Community Health Solutions Fund – COVID-19 Research Grant, Michael Garron Hospital,

Toronto, Canada

- The Ottawa Hospital Foundation COVID-19 Emergency Response Fund, Ottawa, Canada

- International Network of Venous Thromboembolism Clinical Research Networks (INVENT) Kickstarter

Award

- Science Foundation Ireland, Enterprise Ireland, IDA Ireland COVID-19 Rapid Response Funding Call

20/COV/0157

- Southeastern Ontario Academic Medical Organization (SEAMO) COVID-19 Innovation Fund

- P20 GM135007 from the National Institute of General Medical Sciences, NIH

- University of Vermont Medical Center Fund Grant

- College of Medicine Research Center, Deanship of Scientific Research, King Saud University, Riyadh,

Saudi Arabia

Research Personnel and Infrastructure in kind provided by:

- Sinai Health Foundation, Sinai Health, Toronto, Ontario

- Trillium Health Partners Pharmacy Team and Institute for Better Health, Trillium Health Partners, Toronto,

Ontario

- University College Dublin Clinical Research Centre, Dublin, Ireland

- Nursing and Medical Staff, Patients of Mater Misericordiae University Hospital, Dublin, Ireland

- Versiti Blood Research Institute, Milwaukee, Wisconsin

- Division of Hematology research fund, University of Calgary

- University of Calgary Clinical Research Fund

- Calgary Health Trust Fund

- CMO office Dr. Ghazala Belal Balhaj, AlAin Hospital, AlAin, UAE

- Dr. Ghanem AlHassani, Research committee at SEHA Institute, Abu Dhabi, UAE

- Research committee at the Department of Health (DOH), Abu Dhabi, UAE

- Shawn Rhind PhD and Henry Peng PhD, Defence Research Development Canada, Toronto, Canada

- College of Medicine Research Center, King Saud University, Riyadh, Saudi Arabia

- Clinical Trials Unit, King Saud University Medical City, Riyadh, Saudi Arabia

8

3. Supplementary Methods

3.1 Eligibility Criteria

The inclusion criteria:

1) Laboratory confirmed COVID-19 (diagnosis of SARS-CoV-2 via reverse transcriptase polymerase chain reaction as per the World Health Organization protocol or by nucleic acid based isothermal amplification).

Positive test prior to hospital admission OR within first 5 days (i.e. 120 hours) after hospital admission;

2) Admitted to hospital for COVID-19;

3) One D-dimer value above ULN (within 5 days (i.e. 120 hours) of hospital admission) AND EITHER:

a. D-Dimer ≥2 times ULN OR

b. D-Dimer above ULN and Oxygen saturation ≤ 93% on room air;

4) > 18 years of age;

5) Informed consent from the patient (or legally authorized substitute decision maker).

The exclusion criteria:

1) pregnancy;

2) hemoglobin <80 g/L in the last 72 hours; 3) platelet count <50 x 109/L in the last 72 hours;

4) known fibrinogen <1.5 g/L (if testing deemed clinically indicated by the treating physician prior to the

initiation of anticoagulation);

5) known INR >1.8 (if testing deemed clinically indicated by the treating physician prior to the initiation of

anticoagulation);

6) patient already on intermediate dosing of LMWH that cannot be changed (determination of what

constitutes an intermediate dose is to be at the discretion of the treating clinician taking the local

institutional thromboprophylaxis protocol for high risk patients into consideration);

7) patient already on therapeutic anticoagulation at the time of screening (low or high dose nomogram UFH,

LMWH, warfarin, direct oral anticoagulant (any dose of dabigatran, apixaban, rivaroxaban, edoxaban);

8) patient on dual antiplatelet therapy, when one of the agents cannot be stopped safely; 9) known bleeding within the last 30 days requiring emergency room presentation or hospitalization;

10) known history of a bleeding disorder of an inherited or active acquired bleeding disorder;

11) known history of heparin-induced thrombocytopenia;

12) known allergy to UFH or LMWH;

13) admitted to the intensive care unit at the time of screening;

14) treated with non-invasive positive pressure ventilation or invasive mechanical ventilation at the time of

screening (of note: high flow oxygen delivery via nasal cannula is acceptable and is not an exclusion

criterion);

15) Imminent death according to the judgement of the most responsible physician;

16) enrollment in another clinical trial of antithrombotic therapy involving pre-intensive care unit hospitalized

patients.

3.2 Description of Therapeutic Heparin vs. Prophylactic Heparin

Therapeutic Heparin

Therapeutic anticoagulation with low molecular weight heparin (LMWH) or unfractionated heparin (UFH). The

choice of LMWH versus UFH was at the clinician’s discretion and dependent on local institutional supply. LMWH

options are described in Table 1 . UFH was administered using a weight-based nomogram (bolus plus continuous

infusion) with activated partial thromboplastin time (aPTT) or UFH anti-Xa titration according to the center-specific

institutional protocols as per venous thromboembolism treatment (i.e. high dose nomogram). UFH anti-Xa titration was preferred over aPTT if available as achieving a therapeutic aPTT may be challenging in patients with a pro-

inflammatory state such as COVID-19. Therapeutic heparin was administered until hospital discharge, death, day 28

or study withdrawal. If the patient was admitted to the intensive care unit (ICU) or required mechanical ventilatory

support (i.e. patient reached a component of the primary composite outcome), continuation of the allocated

treatment was recommended, as long as the treating physician was in agreement.

Table 1. Therapeutic Heparin

Any of the following strategies could have been used for therapeutic anticoagulation:

CrCl BMI Enoxaparin Dalteparin Tinzaparin UFH

9

≥30

<40 1 mg/kg SC q12h OR

1.5 SC mg/kg q24h

200 units/kg SC q24h

OR

100 IU/kg SC q12h

175 U/kg SC

q24h IV bolus, with continuous infusion to titrate to

institution specific anti-

Xa or aPTT values* ≥40 1 mg/kg q12h& 100 units/kg SC q12h& 175 U/kg SC

daily&

<30 <40

UFH IV bolus, with continuous infusion to titrate to institution specific anti-Xa or aPTT

values* or LMWH per hospital protocol taking BMI into consideration as above

≥40

Abbreviations: CrCl = creatinine clearance; BMI = body mass index; * Initial bolus dose determined by sites,

encouraging use of dosing algorithm designed for treatment of venous thromboembolism. UFH anti‐Xa titration was

preferred over aPTT if available as achieving a therapeutic aPTT may be challenging in patients with a pro‐

inflammatory state such as COVID‐19 &For patients with BMI above 40, measurement of anti-Xa to confirm therapeutic effect was suggested.

Prophylactic Heparin Administration of LMWH, UFH or fondaparinux at thromboprophylactic doses for acutely ill hospitalized medical

patients, in the absence of contraindication, is generally considered standard care. The doses of thromboprophylaxis

only included those listed below (Table 2).

Any of the following strategies could be used for prophylactic heparin doses above those listed was not considered

as prophylactic. A lower dose of either LMWH or UFH than listed below was considered acceptable if due to

extremely low weight/BMI, and considered as part of prophylactic heparin:

Table 2. Prophylactic Heparin

CrCl BMI Enoxaparin Dalteparin Tinzaparin Fondaparinux Unfractionated

Heparin (UFH)

≥30

<40 40 mg SC q24h 5000 units SC

q24h 4500 U SC q24h 2.5 mg SC q24h

5000 units SC q8-

12h

≥40 40 mg SC q12h 5000 units SC

q12h 9000 (+/- 1000)

U SC q24h not applicable

7500 units SC q8h

<30 <40 UFH 5000 units SC q8-12h or LMWH per hospital protocol taking BMI into consideration as above

≥40 UFH 7500 units SC q8h or LMWH per hospital protocol taking BMI into consideration as above

Abbreviations: CrCl = creatinine clearance; BMI = body mass index

Full therapeutic dose anticoagulation (therapeutic dose UFH or LMWH) was permitted as rescue therapy in the

event of suspected or confirmed thromboembolism.

3.3 Primary Outcome

Pre-specified primary composite outcome: ICU admission, non-invasive positive pressure ventilation, invasive

mechanical ventilation or death at 28 days.

If a patient was discharged alive before 28 days, vital status was determined using a telephone follow-up. If a patient

was discharged alive on mechanical ventilation (invasive or non-invasive) prior to day 28, a call to the patient or a

doctor/nurse from the rehabilitation health facility was made to confirm ventilation status on day 28 and their last

day of mechanical ventilation.

Pre-specified secondary outcomes, evaluated up to day 28, included:

1) All-cause death;

2) Composite of ICU admission or all-cause death; 3) Composite of mechanical ventilation or all-cause death;

4) Major bleeding as defined by the ISTH Scientific and Standardization Committee (ISTH-SSC)

recommendation;1

5) Red blood cell transfusion (>1 unit);

10

6) Transfusion of platelets, frozen plasma, prothrombin complex concentrate, cryoprecipitate and/or

fibrinogen concentrate;

7) Renal replacement therapy defined as continuous renal replacement therapy {CRRT} or intermittent

hemodialysis {IHD};

8) Hospital-free days alive; 9) ICU-free days alive;

10) Ventilator-free days alive;

11) Organ support-free days alive;

12) Venous thromboembolism (defined as symptomatic or incidental, suspected or confirmed via diagnostic

imaging and/or electrocardiogram where appropriate recognizing that access to diagnostic imaging may

have been limited due to the COVID-19 pandemic, however confirmatory testing was encouraged);

13) Arterial thromboembolism (defined as suspected or confirmed via diagnostic imaging and/or

electrocardiogram where appropriate recognizing that access to diagnostic imaging may have been limited

due to the COVID-19 pandemic, however confirmatory testing was encouraged);

14) Heparin induced thrombocytopenia;

15) D-dimer at day 2+/- 24 hours.

In addition, the following pre-specified components of the primary composite outcome were pre-specified in the

statistical analysis plan, but not in the protocol:

16) Proportion of subjects with ICU admission;

17) Proportion of subjects with the composite of invasive or non-invasive mechanical ventilation;

18) Proportion of subjects with invasive mechanical ventilation.

Outcome measures were obtained from participants’ hospital medical records and where applicable through a

telephone follow-up. The use of bi-level positive airway pressure (BIPAP) or continuous positive airway pressure

(CPAP) at night or when sleeping for sleep apnea was not considered non-invasive mechanical ventilation or organ

support for the purpose of this trial.

3.4 Pre-specified Outcome Definition for Organ-support Free Days

Defined as the number of days that a patient was alive and free of organ support through 28 days after trial entry.

Organ support was defined as receipt of non-invasive mechanical ventilation, high flow nasal cannula oxygen,

invasive mechanical ventilation, or vasopressor therapy.

• Non-invasive mechanical ventilation was defined as BIPAP or CPAP when used for acute respiratory

support.

• High Flow Nasal Cannula Oxygen was defined as receiving ≥ 30 l/min flow at FiO2 ≥ 40%.

• Invasive mechanical ventilation was defined as positive pressure ventilation through endotracheal tube or

tracheostomy.

• Vasopressor support included the infusion of any vasoactive or inotropic medication.

• A patient must have been extubated and not receiving mechanical ventilation for at least 2 days before

being considered free of mechanical ventilation. If a patient was extubated and re-intubated and placed

back on mechanical ventilation within 1 or 2 days, the patient was considered to be on mechanical

ventilation during those 1 or 2 days before re-intubation.

• Any patient who died during the acute hospital stay was assigned 28 Day Organ-Support Free Days of –1.

• If there was intervening time in which a patient was free of organ support, but went back on organ support,

the intervening time did not count toward the organ support free days endpoint. Only time before organ

support and after the last use of organ support was counted as “free days”.

• If a patient was discharged alive without mechanical ventilation prior to Day 28, the patient was assumed to

be free of organ support after hospital discharge for the remainder of the 28 days.

• If a patient was discharged alive on mechanical ventilation (invasive or non-invasive) prior to day 28, a call

to the patient or a doctor/nurse from the rehabilitation health facility was made to confirm ventilation status

on day 28 and their last day of mechanical ventilation.

3.5 Sample Size Considerations

462 patients (231 per group) were needed to detect a 15% risk difference, from 50% in the control group to 35% in

the experimental group, with power of 90% at a two-sided alpha of 0.048.2–4 No attrition was expected. This

calculation took two interim analyses into account. There was no inflation to account for losses to follow-up because

we expected these to be very infrequent, and given the nature of the trial, included patients, and outcomes, we

11

concluded an absence of the primary outcome in patients discharged alive from hospital before 28 days with missing

outcome data at day 28.

3.6 Extended Description of Statistical Methods

Statistical Analysis

Primary analyses were by the intention-to-treat population of all randomised patients in accordance with the

allocated intervention. We conducted a chi-square test to derive a two-sided p-value for the main analysis of the

primary outcome. We used logistic regression to derive odds ratios with 95% confidence intervals. In addition, we

derived differences in proportions and 95% confidence interval from logistic regression using the observed risk of the primary outcome in the control group,5 and from a binomial model with identity link.

We conducted subgroup analyses accompanied by tests of interaction for the following variables: age, sex, BMI,

time from COVID-19 symptom onset, diabetes mellitus, coronary artery disease, hypertension and race/ethnicity.

Logistic regression and linear regression were used to analyse binary and continuous secondary outcomes after

adjustment for age (used for stratification of randomization).

Secondary outcomes were exploratory and were not adjusted for multiple comparisons. A per-protocol analysis of

the primary outcome was restricted to the per-protocol population of participants, defined as those who received

experimental or control intervention as allocated during the first 48 hours after randomization.

If an outcome was missing in more than 5% of the patients, in addition to the pre-planned strategy of assuming no

outcome if patients were discharged alive from hospital before 28 days, a complete case analysis, an inverse

probability weighted analysis and multiple imputation on outcome was also conducted.

The statistical analysis plan was finalized prior to study closure without prior inspection of the data. All analyses

were conducted in R version 3.6.2 and/or Stata version 15.1, or higher.

Interim Analysis

Interim analyses were done when approximately 25% and 75% of the originally planned number of participants

reached determination of the primary outcome. A group sequential design was employed that applied a one-sided

boundary. The boundary was based on a Hwang-Shih-DeCani spending function for efficacy. When approximately

75% of the originally planned number of participants reached determination of the primary endpoint, we performed a conditional power analysis.

If the conditional power given the accumulated data was <30% and there was robust evidence of harm – either a

relevant increase in the risk of major bleeding in the experimental group and the lower limit of the 95% confidence

interval for major bleeding excluded 5% on an absolute risk difference scale; or a relevant increase in the risk of all-

cause death in the experimental group and the lower limit of the 95% confidence interval for death excluded 1% on

an absolute risk difference scale - the protocol called for a non-binding recommendation to stop the trial by the Data

and Safety Monitoring Board (DSMB). If the conditional power was <30%, but there was no robust evidence of

harm, the protocol called for completing recruitment as planned. The rationale for this approach was that prevention

of death (a component of the primary outcome) overrides short-term safety. If major bleeds led to bleeding related

deaths to such an extent that no mortality benefit was likely to be realized, the trial would have been stopped. If the conditional power was ≥30 and <60%, the protocol called for completing recruitment as planned. If the conditional

power was ≥60% and <80%, the protocol called for a non-binding recommendation to increase the sample size to

achieve 80% power, if deemed feasible from a recruitment perspective. If the conditional power was ≥80%, the

protocol called for completing recruitment as planned, provided that the interim analysis against the one-sided

boundary for efficacy was negative.

3.7 Adjudication

Outcomes were independently and blindly adjudicated by two clinical content experts for the English language

source documentation, and two clinical content experts for the Portuguese source documentation from the Brazilian

site.

The adjudicators reviewed de-identified and treatment allocation redacted, source documentation (e.g. clinical notes,

discharge summary, diagnostic imaging, laboratory tests, autopsy reports etc.) to confirm the presence of clinical

events specified in the protocol, and the date of occurrence for the following:

1) ISTH-defined major bleeding

12

2) Heparin induced thrombocytopenia (HIT)

3) Venous thromboembolism

4) Arterial thromboembolism

5) Mechanical ventilation, including whether invasive or non-invasive

6) Intensive care unit admission 7) Death, including cause of death

Each clinical event was reviewed in duplicate by two independent adjudicators who determined whether the event

met the pre-specified criteria (per definitions in the protocol). The events were classified as a “Definite Event",

"Probable Event" or “Not an Event” (see adjudication table from the adjudication manual below). The final adjudication result was based on consensus. If there was disagreement between the two adjudicators the medical

monitors broke the tie. Additional source documentation could be requested.

Adjudication Checklist:

Adjudicator to complete:

Definite Event

Probable Event

Not an Event

If major hemorrhage, indicate which criterion of the ISTH definition was met:

1. Fatal bleeding, and/or

2. Symptomatic bleeding in critical area or organ (such as intracranial, intraspinal, intraocular, retroperitoneal,

intraarticular or pericardial, or intramuscular with compartment syndrome, and/or

3. Bleeding causing a fall of hemoglobin level of 20g/L or more, or leading to transfusion of two or more units

of whole blood or red cells.

If HIT, indicate if laboratory test proven Yes / No and if accompanied by a thrombotic event

Yes / No

If ATE, indicate if suspected ATE OR diagnostically confirmed ATE

If ATE, indicate type: ischemic stroke OR MI, OR limb ischemia OR other _____________ _____________________________________________________________________________________

If VTE, indicate if suspected VTE OR diagnostically confirmed VTE

If VTE, indicate if symptomatic VTE OR asymptomatic VTE OR unclear

If VTE, indicate type: PE OR DVT OR splanchnic VT OR cerebral VT OR other ________

_____________________________________________________________________________________

If PE: indicate if segmental/beyond OR subsegmental

If DVT: indicate if proximal (above knee) OR distal (below knee)

If ICU admission, Yes, patient was admitted OR No, patient was not admitted OR unclear

If yes, indicate rationale for transfer to ICU __________________________________________

_____________________________________________________________________________ OR unclear

If mechanically ventilated, indicate if non-invasive OR invasive

If patient died, indicate cause of death: _____________________________________ OR unclear

Adjudicator signature + date: _____________________________

13

Adjudication Manual

EVENT TYPE INFORMATION NEEDED DOCUMENTS REQUIRED

(EXAMPLES)

Death □ Date of death

□ Cause of death

Medical notes, death certificate,

discharge summary

ICU admission □ Date of transfer to ICU

□ Rationale for transfer

Medical notes, medical orders

Invasive Mechanical

ventilation □ Stat date of invasive mechanical ventilation (i.e.

endotracheal intubation with mechanical ventilation)

□ Rationale for invasive mechanical ventilation (e.g.

hypoxemic respiratory failure, airway protection due

to compromised neurologic status)

□ Max FiO2 (e.g. 0.80 or 80%)

□ Stop date (if applicable)

Medical notes, medical orders,

respiratory therapy notes

Diagnostic imaging: Chest Xray

indicating placement of endotracheal

tube

Noninvasive

Mechanical

Ventilation (Positive

pressure ventilation)

□ Start date

□ BIPAP or CPAP

□ Rationale for noninvasive mechanical ventilation (e.g.

hypoxemic respiratory failure)

□ Max FiO2 (e.g. 0.80 or 80%)


Medical notes, medical orders,

respiratory therapy notes, nursing notes

Major bleeding* □ Date of onset

□ Date of resolution (if applicable)

□ Transfusion (# of red cell units and date transfused)

□ Location of bleed in critical area or organ (e.g.

intracranial, intraspinal, intraocular, retroperitoneal,

intraarticular, pericardial, or intramuscular with

compartment syndrome)

□ Fatal bleeding (yes or no)

□ Fall in hemoglobin by ≥ 20 g/L

Medical notes, medical orders, nursing

notes (e.g. transfusion order/nursing

documentation of transfusion administration),surgical note (if

applicable)

Lab result: fall in hemoglobin by >/=

20 g/L (if applicable)

Diagnostic imaging reports (if

available): CT, MRI, Ultrasound report

Heparin-induced

thrombocytopenia

□ Date of onset


□ Laboratory confirmation of heparin-induced

thrombocytopenia

□ Evidence of secondary thromboembolism (if

applicable)

□ Treatment given (medication order)


Lab results:

- 5-day trend of platelet count AND

- Immunologic based assay (ELISA

or LIA) assay evaluating for

heparin-PF4 antibodies and/or

Serotonin release assay

Diagnostic imaging (per venous

thromboembolism and arterial

thromboembolism categories below if

patient experienced secondary

thromboembolism)

14

Venous

thromboembolism □ Date of onset


□ Type (PE, DVT, splanchnic vein thrombosis, cerebral

vein thrombosis, or other)

o If PE: segmental/beyond or subsegmental o If DVT: distal (below knee) or proximal (above

knee)

□ Symptomatic or asymptomatic

□ Suspected or confirmed



Lab result: D-dimer (if available) from

date of onset


available): CT, Doppler Ultrasound, MRI, ventilation/perfusion lung scan

Arterial

thromboembolism □ Date of onset


□ Type (ischaemic stroke, MI, or limb ischemia, other)

□ Suspected or confirmed



Lab results (if available) from date of

onset:

- Troponin

- Lactate (venous or arterial)


available): Doppler ultrasound,

ultrasound, angiogram, ECG, CT, MRI,

echocardiogram

Organ Support via

High-Flow Nasal

Cannula

□ Start date


□ Max O2 flow rate (should be >30 L/min to qualify,

per our trial definition, as HFNC)

□ Max FiO2 value (e.g. 0.80 or 80%)

Medical notes, respiratory therapy

notes, nursing notes

Organ Support via

Vasopressor/Inotropic

therapy

□ Start date


□ Vasopressor/inotrope examples:

▪ Norepinephrine (Levophed, Levo)

▪ Epinephrine (Epi)

▪ Vasopressin (Vaso)

▪ Dopamine (Dop)

▪ Dobutamine (Dobu)

Medical notes, nursing notes, medical

orders

*Major bleeding defined by ISTH Scientific and Standardization Committee (ISTH-SSC) recommendation:

1) Fatal bleeding, and/or 2) symptomatic bleeding in critical area or organ (such as intracranial, intraspinal, intraocular,

retroperitoneal, intraarticular or pericardial, or intramuscular with compartment syndrome, and/or

3) bleeding causing a fall of hemoglobin level of 20g/L or more, or leading to transfusion of two or more

units of whole blood or red cells.

3.8 Data and Safety Monitoring Board

The DSMB acted in an advisory capacity to the principal investigators to monitor participant safety, data quality and

evaluate the progress of the trial. The DSMB was composed of a biostatistician, a hematologist, a general internist

and an intensive care specialist. The four members were not study investigators. The DSMB convened meetings at

the formal interim analyses mentioned above (section 3.6), and also when approximately 10% and 50% of the

originally planned number of participants reached determination of the primary outcome. When 10%, 25% and 50%

of the originally planned number of participants reached determination of the primary outcome, this

recommendation was at the discretion of the DSMB. When approximately 75% of the originally planned number of

participants reached determination of the primary outcome, the DSMB was required to also take into account the

conditional power when doing the safety review. The recommendations were made by a formal majority vote based

on safety concerns as evidenced by statistical and clinical judgment, progress of the trial including data quality and

15

accrual/retention, and new scientific or therapeutic developments that may have an impact on the safety. The DSMB

was immediately informed of any serious adverse events (SAEs) which were potentially study drug related.

Moreover, the DSMB chair was notified within 24 hours of any major bleed or occurrence of heparin induced

thrombocytopenia. The Data Management and Coordination Center (DMCC) was responsible for the data analysis

and the DMCC statistician provided the interface with DSMB members. The DSMB received reports of enrollment and events, including events as reported by the site and had full access to the data.

3.9 Trial Administration

Protocols in English (NCT04362085) and Portuguese (NCT04444700) were harmonized in all aspects, except for an initial material difference in eligibility criteria: between July 3, 2020, and October 22, 2020, patients in the single

Brazilian site of the trial were eligible if they had normal D-Dimer levels, but an oxygen saturation of ≤93% on

room air. This led to the initial inclusion of 11 patients in Brazil with normal D-Dimer levels at baseline. These

patients were included in the intention-to-treat analysis. From October 23, 2020 onwards, protocols were fully

harmonized, requiring elevated D-Dimer levels at baseline in all patients. Data from all sites, including the site in

Brazil, were entered in a common database and managed centrally by the trial’s data coordination centre (Applied

Health Research Centre, St. Michael’s Hospital, University of Toronto) according to a single set of standard

operating procedures.

The original protocol of RAPID BRAZIL initially approved on June 17, 2020 by the Institutional Research Ethics

Board and Brazilian National Research Ethics Commission is version 3.0. All subsequent versions were anchored from version 3.0. The latest version of the protocol is version 5.2 approved on April 13, 2021.

16

4. Supplementary Figures and Tables

Figure S1. Trial Schematic

aAdministered until hospital discharge, death or day 28, if the patient is admitted to the ICU or required ventilatory support, we recommended continuation of the allocated treatment as long as the treating physician was in agreement. bA single D-dimer test (if not collected through standard of care) on day 2 after randomization (±24 hours) was collected for participants in both study arms (considering the day of randomization as day 1).

Eligibility

Randomization (1:1)

Prophylactic Heparina

Therapeutic Heparina

The primary composite outcome of ICU admission, non-invasive positive pressure

ventilation, invasive mechanical ventilation or death at 28 days.

One D-dimer after randomizationb

One D-dimer after randomizationb

17

Table S1. Recruitment Numbers By Site

Therapeutic

Heparin Prophylactic

Heparin

Country and Name of Site (N=228) (N=237)

no. of patients (%)

Brazil

Hospital das Clinicas HCFMUSP 54 (23.7) 51 (21.5)

Canada Hôpital Charles Lemoyne 15 (6.6) 18 (7.6)

William Osler Health Systema 15 (6.6) 15 (6.3)

Mount Sinai Hospital 7 (3.1) 9 (3.8)

Trillium Health Partnersa 7 (3.1) 7 (3.0)

St. Joseph's Health Centre Toronto 7 (3.1) 7 (3.0)

Alberta Health Servicesa 7 (3.1) 7 (3.0)

The Ottawa Hospitala 5 (2.2) 5 (2.1)

Hôpital du Sacré-Coeur de Montréal 2 (0.9) 4 (1.7)

Michael Garron Hospital 3 (1.3) 2 (0.8)

St. Michael's Hospital 3 (1.3) 1 (0.4)

University of Alberta Hospital 0 (0.0) 2 (0.8)

Southlake Regional Health Centre 1 (0.4) 0 (0.0)

Hospital Maisonneuve-Rosemont 0 (0.0) 1 (0.4)

Queen Elizabeth II Health Sciences Centre 0 (0.0) 0 (0.0)

Ireland Mater Misericordiae University Hospital 11 (4.8) 12 (5.1)

Saudi Arabia King Saud University Medical City 29 (12.7) 31 (13.1)

King Fahad Medical City 28 (12.3) 28 (11.8)

King Faisal Specialist Hospital 14 (6.1) 17 (7.2)

United Arab Emirates Al Ain Hospital 7 (3.1) 6 (2.5)

United States of America Versiti 12 (5.3) 13 (5.5)

University of Vermont Medical Center 1 (0.4) 1 (0.4)

Barnes Jewish Hospital 0 (0.0) 0 (0.0) aSome RAPID trial sites include multiple hospitals. There were 28 total recruiting sites.

18

Table S2. Means and Medians and Measures of Dispersion of Continuous Baseline Characteristics Therapeutic Heparin Prophylactic Heparin

Characteristic (N=228) (N=237)

Age (years)

Mean (SD) 60.4 (14.1) 59.6 (15.5)

Median (IQR) 60.0 (51.8, 70.0) 60.0 (49.0, 70.0)

Body mass index (kg/m2)a

Mean (SD) 30.3 (6.4) 30.2 (7.0)

Median (IQR) 28.8 (25.8, 33.1) 29.3 (25.8, 33.0)

Duration of symptoms prior to hospitalization (days)b

Mean (SD) 7.1 (5.1) 7.1 (5.2)

Median (IQR) 7.0 (3.0, 10.0) 7.0 (3.0, 10.0)

Duration of hospitalization before randomization (days)

Mean (SD) 1.5 (1.1) 1.4 (1.0)

Median (IQR) 1.0 (1.0, 2.0) 1.0 (1.0, 2.0)

Platelet count (109/L)c

Mean (SD) 233.7 (95.7) 237.8 (95.3)

Median (IQR) 220.5 (169.0, 276.0) 221.0 (171.0, 288.0)

Creatinine (μmol/L)d

Mean (SD) 84.6 (44.1) 85.9 (58.2)

Median (IQR) 77.0 (64.0, 91.0) 75.1 (62.2, 94.6)

D-dimer ratio (D-dimer x ULN)

Geometric mean (SD)e 2.1 (0.7) 2.5 (0.9)

Median (IQR) 2.0 (1.4, 2.9) 2.0 (1.4, 3.4) Abbreviations: IQR, interquartile range; SD, standard deviation; ULN, upper limit of normal. aBody-mass index (BMI) is the weight in kilograms divided by the square of the height in meters; Data regarding BMI was missing for 6 participants in the therapeutic heparin group and 4 participants in the prophylactic heparin group. bData regarding duration of symptoms prior to hospitalization were missing for 1 patient in the therapeutic heparin group and 5 for the

prophylactic heparin group. cData regarding platelet count was missing for 16 patients in the therapeutic heparin group and 24 patients in the prophylact ic heparin group. dData regarding creatinine was missing for 14 patients in the therapeutic heparin group and 23 patients in the prophylactic heparin group. eSD for the natural logarithm of D-dimer ratios (D-dimer levels x ULN).

19

Table S3. Duration and Type of Heparin Used

Therapeutic Heparin Prophylactic Heparin

Heparin (N=228) (N=237)

no. of patients (%)

Duration of anticoagulation (days)a

Mean (SD) 6.5 (5.4) 6.3 (5.4)

Median (IQR) 6.0 (3.0, 8.0) 5.0 (3.0, 8.0)

Dalteparin 25 (11.0) 25 (10.5)

Enoxaparin 188 (82.5) 183 (77.2)

Fondaparinux 0 (0.0) 1 (0.4)

Tinzaparin 11 (4.8) 13 (5.5)

Unfractionated heparin 3 (1.3) 14 (5.9)

Abbreviations: SD, standard deviation. IQR. interquartile range.

aData on the duration of anticoagulation were missing for 4 patients in the therapeutic heparin group and 5 patients in the prophylactic heparin group.

20

Table S4. Concomitant Treatments Received


No Death (N=224) Death (N=4) No Death (N=219) Death (N=18)

no. of patients (%)

Systemic Corticosteroid 172 (76.8) 3 (75.0) 167 (76.3) 14 (77.8)

Remdesivir 30 (13.4) 1 (25.0) 27 (12.3) 3 (16.7)

Tocilizumab 10 (4.5) 0 (0.0) 11 (5.0) 2 (11.1) Treatments received over course of study duration, pre- and post-randomization combined. See Table 1 for medication use at baseline.

21

Table S5. Means and Medians and Measures of Dispersion of Continuous Outcomes

Outcome

Therapeutic Heparin

Prophylactic Heparin

Odds Ratio or Ratio of Geometric

Means (95% CI)

P value

(N=228) (N=237)

Ventilator-free days alive (days)

Mean (SD) 26.5 (5.6) 24.7 (8.5) 1.77 (1.02, 3.08) 0.042

Median (IQR) 28.0 (28.0, 28.0) 28.0 (28.0, 28.0)

Organ support-free days alive (days)

Mean (SD) 25.8 (6.2) 24.1 (8.8) 1.41 (0.90, 2.21) 0.13

Median (IQR) 28.0 (28.0, 28.0) 28.0 (28.0, 28.0)

ICU-free days alive (days)

Mean (SD) 26.0 (6.1) 24.2 (8.8) 1.51 (0.94, 2.41) 0.087

Median (IQR) 28.0 (28.0, 28.0) 28.0 (28.0, 28.0)

Hospital-free days alive (days)

Mean (SD) 19.8 (7.3) 18.4 (9.2) 1.09 (0.79, 1.50) 0.59

Median (IQR) 22.0 (18.0, 25.0) 22.0 (18.0, 25.0)

D-dimer ratio (D-dimer x ULN)

Geometric mean (SD)a 1.9 (0.7) 2.4 (0.9) 0.88 (0.78, 0.99) 0.032

Median (IQR) 1.8 (1.3, 2.6) 2.0 (1.2, 3.6)

Abbreviations: IQR, interquartile range; SD, standard deviation; ULN, upper limit of normal. aRatio of geometric means of D-dimer ratios (D-dimer x ULN) of day 2±24h post-randomization, adjusted for baseline geometric means of D-

dimer ratios using analysis of covariance. SD for the natural logarithm of D-dimer ratios. The day 2±24 hours D-dimer was missing for 66 in the therapeutic heparin group and 64 in the prophylactic heparin group.

22

Table S6. Primary Cause of Death


Cause (N=228) (N=237)

no. of patients (%)

Hypoxemic respiratory failure 4 (100.0) 13 (72.2)

Multi-system organ failure 0 (0.0) 5 (27.8)

There were no cases of sudden, unexplained death.

23

Figure S2. Time-to-event Analyses of the Primary Composite Outcome

The hazard ratio comparing therapeutic with prophylactic heparin was 0.72 (95% confidence interval, 0.47 to 1.10).

Abbreviations: TH, therapeutic heparin. PH, prophylactic heparin. Primary composite outcome defined as death, invasive mechanical ventilation, non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation or ICU admission. Cumulative primary composite outcome curves through 28 days in the two study groups.

0.0%

5.0%

10.0%

15.0%

20.0%

0 7 14 21 28

Time (days)

Cu

mu

lative

Pro

ba

bili

tyTH PH

Time to Primary Compostite Outcome

228 197 193 192 192237 197 187 186 186PH

TH

0 7 14 21 28

Time (days)

Number at risk

24

Figure S3. Time-to-event Analyses of All-cause Death

The hazard ratio comparing therapeutic with prophylactic heparin was 0.23 (95% confidence interval, 0.08 to 0.66). Abbreviations: TH, therapeutic heparin. PH, prophylactic heparin. Cumulative all-cause death curves through 28 days in the two study groups.

0.0%

2.0%

4.0%

6.0%

0 7 14 21 28

Time (days)

Cu

mu

lative

Pro

ba

bili

tyTH PH

Time to Death

228 228 224 224 224237 232 226 221 219PH

TH

0 7 14 21 28

Time (days)

Number at risk

25

Figure S4. Time-to-event Analyses of Mechanical Ventilation

The hazard ratio comparing therapeutic with prophylactic heparin 0.83 (95% confidence interval, 0.47 to 1.48).

Abbreviations: TH, therapeutic heparin. PH, prophylactic heparin. Cumulative invasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation curves through 28 days in the two study groups.

0.0%

3.0%

6.0%

9.0%

0 7 14 21 28

Time (days)

Cu

mu

lative

Pro

ba

bili

tyTH PH

Time to Mechanical Ventilation

228 209 207 206 205237 212 203 199 199PH

TH

0 7 14 21 28

Time (days)

Number at risk

26

Figure S5. Time-to-event Analyses of Intensive Care Unit Admission

The hazard ratio comparing therapeutic with prophylactic heparin 0.81 (95% confidence interval, 0.51 to 1.28). Abbreviations: TH, therapeutic heparin. PH, prophylactic heparin. Cumulative ICU admission curves through 28 days in the two

study groups.

0.0%

5.0%

10.0%

15.0%

0 7 14 21 28

Time (days)

Cu

mu

lative

Pro

ba

bili

tyTH PH

Time to ICU Admission

228 197 192 192 192237 200 188 187 187PH

TH

0 7 14 21 28

Time (days)

Number at risk

27

Table S7. Thromboembolism


Event (N=228) (N=237) no. of patients (%)

Venous Deep vein thrombosisa 1 (0.4) 1 (0.4) Pulmonary embolismb 1 (0.4) 5 (2.1)

Arterial Myocardial infarction 0 (0.0) 1 (0.4)

a1 patient in the therapeutic heparin group (symptomatic, diagnostically confirmed, proximal deep vein), 1 patient in the prophylactic

heparin group (incidental deep venous thrombosis, diagnostically confirmed, proximal deep vein). b1 patient in the therapeutic heparin group (symptomatic, diagnostically confirmed, segmental pulmonary artery or beyond), 5 patients in the prophylactic heparin group (all symptomatic, all diagnostically confirmed, 4 segmental pulmonary artery or beyond, 1

sub-segmental pulmonary artery).

28

Table S8. ISTH Major Bleeding Events

Randomised

treatment allocation

Anticoagulation received 24

hours prior to event

Fatal bleeding

Symptomatic bleeding in

critical area or organa

Bleeding causing a fall of hemoglobin

of 20g/L or more

Bleeding leading to

transfusion of two or more

units of whole blood or red

cells

Relatednessb Concomitant medications

Patient 1

Therapeutic Heparin

Therapeutic dose LMWH

No Intramuscular Yes Yes Not related Systemic

corticosteroid

Patient 2

Therapeutic Heparin

Prophylactic dose LMWH

No No Yes Yes Unlikely

Systemic corticosteroid,

Antiplatelet agent

Patient 3


Intermediate dose LMWH

No No Yes Yes Unlikely Systemic

corticosteroid

Patient 4



No No Yes Yes Unlikely

Systemic corticosteroid,

Antiplatelet agent

Patient 5



No Retroperitoneal Yes Yes Probable Antiplatelet

agent Patient

6 Prophylactic

Heparin None* No No Yes Yes Unlikely

Systemic corticosteroid

Major bleeding defined by the International Society on Thrombosis and Haemostasis (ISTH) Scientific and Standardization Committee.1; LWMH = low molecular weight heparin. aAll non-critical area/organ bleeding events were gastrointestinal in origin. bRelatedness was independently and blindly adjudicated. *This event occurred 11 days post-hospital discharge.

29

Table S9. Bleeding Events by Concomitant Treatments Received


no. of patients (%)

Major Bleeding 2 4

Systemic corticosteroid only 2 (100.0) 2 (50.0)

Antiplatelet agent only 0 (0.0) 1 (25.0)

Antiplatelet and systemic corticosteroid 0 (0.0) 1 (25.0)

No antiplatelet and systemic corticosteroid 0 (0.0) 0 (0.0)

No major bleeding 226 233

Systemic corticosteroid only 125 (55.3) 129 (55.4)

Antiplatelet agent only 8 (3.5) 13 (5.6)

Antiplatelet and systemic corticosteroid 25 (11.1) 24 (10.3)

No antiplatelet and systemic corticosteroid 68 (30.1) 67 (28.8)

Major bleeding defined by the International Society on Thrombosis and Haemostasis Scientific and Standardization Committee.1

30

Figure S6. Post-Hoc Subgroup Analysis of All-Cause Death

Subgroup-specific odds ratios derived from logistic regression. Point estimates are plotted as dark circles; the horizontal lines

represent the 95% confidence intervals. Odds ratio less than 1.0 favors therapeutic heparin. BMI, body mass index in kg/m2.

Subgroup

All Patients

Therapeutic Heparin

Age


<=65

>65

Odds Ratio (95% CI)

Sex

Male

Female

Race/ethnicity

White

Asian

Black or African American

Hispanic or Latino

BMI

<=30

>30

Diabetes mellitus

Yes

No

Coronary artery disease

Yes

No

Hypertension

Yes

No

Duration of symptoms pr ior to hospitalization

<=7 days

>7 days

Region

North America

Brazil

Middle East

Ireland

D−dimer levels

<2x upper limit of normal

>=2x upper limit of normal

Use of systemic cor ticosteroids

Yes

No

4/228 (1.8)

0/144 (0)

4/84 (4.8)

1/123 (0.8)

3/105 (2.9)

2/162 (1.2)

1/27 (3.7)

0/18 (0)

0/14 (0)

4/131 (3.1)

0/91 (0)

0/83 (0)

4/144 (2.8)

1/16 (6.2)

3/201 (1.5)

4/108 (3.7)

0/118 (0)

4/125 (3.2)

0/102 (0)

4/85 (4.7)

0/54 (0)

0/78 (0)

0/11 (0)

2/113 (1.8)

2/115 (1.7)

3/161 (1.9)

1/67 (1.5)

18/237 (7.6)

3/146 (2.1)

15/91 (16.5)

9/141 (6.4)

9/96 (9.4)

14/163 (8.6)

0/38 (0)

0/23 (0)

3/10 (30)

12/133 (9)

5/100 (5)

4/77 (5.2)

14/157 (8.9)

5/18 (27.8)

13/211 (6.2)

14/117 (12)

4/118 (3.4)

13/131 (9.9)

2/101 (2)

13/92 (14.1)

4/51 (7.8)

1/82 (1.2)

0/12 (0)

9/125 (7.2)

9/112 (8)

11/162 (6.8)

7/75 (9.3)

0.22 (0.07 to 0.65)

0.14 (0.01 to 2.77)

0.25 (0.08 to 0.80)

0.12 (0.01 to 0.97)

0.28 (0.07 to 1.09)

0.13 (0.03 to 0.60)

4.36 (0.17 to 111.13)

0.07 (0.00 to 1.63)

0.32 (0.10 to 1.01)

0.10 (0.00 to 1.74)

0.10 (0.00 to 1.85)

0.29 (0.09 to 0.91)

0.17 (0.02 to 1.69)

0.23 (0.06 to 0.82)

0.28 (0.09 to 0.89)

0.11 (0.01 to 2.02)

0.30 (0.10 to 0.95)

0.19 (0.01 to 4.10)

0.30 (0.09 to 0.96)

0.10 (0.00 to 1.85)

0.35 (0.01 to 8.62)

0.23 (0.05 to 1.10)

0.20 (0.04 to 0.96)

0.26 (0.07 to 0.96)

0.15 (0.02 to 1.24)

0.1 1.0 10.0

0.72

0.49

0.12

0.45

0.50

0.83

0.55

0.79

0.77

0.90

0.65

P - value

Therapeutic Better Prophylactic Better

31

Table S10. Per Protocol Analysis of the Primary Outcome and Its Components

Outcome

Therapeutic Heparin Prophylactic Heparin Odds Ratio (95%

CI) (N=216) (N=227)

no. of patients (%)

Primary Composite Outcomea 34 (15.7) 47 (20.7) 0.72 (0.44, 1.17)

Components of the primary composite outcome Death from any cause 4 (1.9) 17 (7.5) 0.23 (0.08, 0.71)

Invasive mechanical ventilation 9 (4.2) 13 (5.7) 0.72 (0.30, 1.71)

Any mechanical ventilationb 18 (8.3) 22 (9.7) 0.85 (0.44, 1.63)

Intensive care unit admission 30 (13.9) 37 (16.3) 0.83 (0.49, 1.40)

Per protocol analysis excluded patients who did not receive their allocated treatment during the first 48 hours after randomization. aDefined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission. bInvasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation.

32

Table S11. Sensitivity Analysis 1 of the Primary Outcome and Its Components

Outcome

Therapeutic Heparin Prophylactic Heparin Odds Ratio (95%

CI) (N=224) (N=230)

no. of patients (%)


Components of the primary composite outcome

Death from any cause 4 (1.8) 18 (7.8) 0.21 (0.07, 0.64)



Intensive care unit admission 33 (14.7) 42 (18.3) 0.77 (0.47, 1.27) Sensitivity analysis 1 excluded patients who did not meet a component of the primary composite outcome and did not have a follow-up up to day 28; 4 patients in therapeutic heparin group and 7 patients in the prophylactic heparin group. aDefined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission. bInvasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation.

33


Outcome

Therapeutic Heparin Prophylactic

Heparin Odds Ratio (95% CI) (N=222) (N=231)

no. of patients (%)






Intensive care unit admission 32 (14.4) 39 (16.9) 0.83 (0.50, 1.38) Sensitivity analysis 2 excluded those who did not satisfy all eligibility criteria (i.e. those with a negative d-dimer; 6 patients in the therapeutic heparin group and 5 in the prophylactic heparin group). aDefined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission. bInvasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation.

34


Outcome


Odds Ratio (95% CI) (N=218) (N=224)

no. of patients (%)

Primary Composite Outcomea 36 (16.5) 48 (21.4) 0.73 (0.45, 1.17) Components of the primary composite outcome




Intensive care unit admission 32 (14.7) 39 (17.4) 0.82 (0.49, 1.36) Sensitivity analysis 3 excluded patients who did not meet a component of the primary composite outcome, did not have a follow-up up to day

28 and those who did not satisfy all eligibility criteria; 10 patients in the therapeutic heparin group and 13 patients in the prophylactic heparin group. aDefined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission. bInvasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation.

35

Table S14. Intention-to-Treat Analysis of the Primary Outcome and Its Components Adjusted for Age

Outcome

Therapeutic Heparin Prophylactic

Heparin Odds Ratio (95% CI)

(N=228) (N=237)

no. of patients (%)






Intensive care unit admission 33 (14.5) 42 (17.7) 0.78 (0.47, 1.29)

Intention-to-treat analysis of the primary outcome and its components adjusted for age taking into account that randomization was stratified by age. aDefined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission. bInvasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation.

36

Table S15. Intention-to-Treat Analysis of the Primary Outcome with Time-by-Treatment Interaction

Analysis


Odds Ratio (95% CI) (N=228) (N=237)

no. of patients (%)

Primary Analysis 37 (16.2) 52 (21.9) 0.69 (0.43, 1.10)

Analysis adjusted for time 37 (16.2) 52 (21.9) 0.69 (0.43, 1.10)

Intention-to-treat analysis of the primary outcome according to primary analysis and adjusted for time, including a time-by-treatment interaction. Primary outcome defined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission.

To address changes in co-interventions over time due to emerging evidence from Covid-19 clinical trials, a logistic regression model was used to fit a time by treatment interaction where time was days since first randomised subject. Time was modelled with a restricted cubic spline having 3 knots. Three knots were chosen because of the modest number of events. The model with splines and interactions revealed little evidence for an interaction (p = 0.85) or non-linearity (p = 0.95). Given these results a linear additive model was fit to estimate the time adjusted treatment effect. In this model there was strong evidence of a time effect (p=0.0086) while the evidence for a treatment effect was identical to the unadjusted analysis (p for treatment effect=0.12).

37

Table S16. Intention-to-Treat Analysis of the Primary Outcome Estimating Risk Differences

Analysis

Therapeutic Heparin

Prophylactic Heparin Risk Difference

(95% CI) (N=228) (N=237)

no. of patients (%)

Estimated from logistic regression 37 (16.2) 52 (21.9) -5.7% (-11.2%, 1.7%)

Estimated from binomial model 37 (16.2) 52 (21.9) -5.7% (-12.9%, 1.4%)

Intention-to-treat analysis of the primary outcome estimating risk differences from logistic regression and a binomial model with identity link.

Primary outcome defined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission.

The primary outcome was reanalysed with a binary model and identity link to estimate the absolute risk difference. This analysis yielded nearly identical results to the logistic regression and risk difference estimated from that model. The evidence for a treatment effect was similar to the evidence based on logistic regression (p for treatment effect=0.12).

38

Table S17. Sensitivity Analyses of D-Dimer Levels at Day 2

Analysis Ratio of Geometric Means

(95% CI)

Primary analysis (complete case) 0.88 (0.78, 0.99)

Inverse probability weighted analysis 0.87 (0.78, 0.98)

Multiple Imputation 0.91 (0.81,1.03) D-dimer levels at day 2±24 hours post-randomization were missing for 66 (29.0%) in the therapeutic heparin group and 64 (27.0%) in the

prophylactic heparin groups. As pre-specified, we therefore used an inverse probability weighted analysis and multiple imputation to derive ratios of geometric means. Ratio of geometric means of D-dimer level x ULN of day 2±24h post-randomization, adjusted for baseline geometric means of D-dimer

levels x ULN using analysis of covariance. SD for the natural logarithm of D-dimer levels x ULN.

39

Table S18. Study Drug Not Received as Allocated within the First 48 hours Therapeutic Heparin Prophylactic Heparin (N=228) (N=237)

no. of patients (%)

Any change 6 (2.6) 5 (2.1) Discontinued 2 (0.9) 1 (0.4) Prophylactic 2 (0.9) - Intermediate 2 (0.9) 1 (0.4)

Therapeutic - 3 (1.3) Study drug not received as allocated defined as not received as allocated within the first 48 hours post randomization or changed without clear clinical indication.

40

Table S19. Study Drug Change within the First 48 hours With and Without Clear Clinical Indication Therapeutic Heparin Prophylactic Heparin

(N=228) (N=237)

no. of patients (%)

Clear clinical indication

Suspected or Confirmed Thromboembolism 0 (0.0) 5 (2.1)b

Change in Creatinine Clearance 1 (0.4)a 0 (0.0)

Bleeding 1 (0.4)a 1 (0.4)a

Intensive care unit admission 2 (0.9)a 0 (0.0)

Palliative care 1 (0.4)a

No clear clinical indication

Clinician discretion 4 (1.8)a 5 (2.1)c

Patient refusal 2 (0.9)a 0 (0.0) Heparin dose change lasting over 24 hours within the first 48 hours of study treatment period. aHeparin dose decreased or heparin discontinued. bHeparin dose increased. c3 patients received therapeutic dose heparin, 1 received intermediate dose heparin, 1 had

their heparin discontinued.

41

Table S20. Primary and Secondary Outcomes until Hospital Discharge

Outcome

Therapeutic Heparin


Odds Ratio or Ratio of

Geometric Means (95% CI)

(N=228) (N=237) p value

no. of patients (%)

Primary composite outcomea 36 (15.8) 51 (21.5) 0.68 (0.43, 1.10) 0.11


Death from any cause 4 (1.8) 17 (7.2) 0.23 (0.08, 0.70) 0.009

Invasive mechanical ventilation 10 (4.4) 15 (6.3) 0.68 (0.30, 1.55) 0.36

Any mechanical ventilationb 20 (8.8) 25 (10.5) 0.82 (0.44, 1.52) 0.52

ICU admission 32 (14.0) 41 (17.3) 0.78 (0.47, 1.29) 0.33

Death or any mechanical ventilation 22 (9.6) 37 (15.6) 0.58 (0.33, 1.01) 0.056

Death or ICU admission 35 (15.4) 49 (20.7) 0.70 (0.43, 1.12) 0.14

Renal replacement therapyc 1 (0.4) 4 (1.7) 0.26 (0.03, 2.33) 0.23

Thromboembolismd

Venous 2 (0.9) 6 (2.5) 0.34 (0.07, 1.71) 0.19

Arterial 0 (0.0) 1 (0.4) - -

Bleeding

ISTH major bleedinge 2 (0.9) 3 (1.3) 0.69 (0.11, 4.19) 1.0

Red blood cell transfusion (>1 unit) 2 (0.9) 8 (3.4) 0.25 (0.05, 1.21) 0.11

Transfusion of other blood components or productsf

0 (0.0) 0 (0.0) - -

Heparin induced thrombocytopenia 0 (0.0) 0 (0.0) - -

Abbreviations: ICU, intensive care unit; CI, confidence interval; SD, standard deviation; ISTH, International Society on Thrombosis and Hemostasis. aDefined as death, invasive mechanical ventilation, non-invasive mechanical ventilation or ICU admission. bInvasive or non-invasive (bilevel or continuous positive airway pressure) mechanical ventilation. cContinuous renal replacement therapy or intermittent hemodialysis.

dAll diagnostically confirmed except for 1 symptomatic deep vein thrombosis in the prophylactic heparin group, which could not be definitively confirmed as diagnostic imaging was not done during acute symptomatic period. eMajor bleeding defined by the ISTH Scientific and Standardization Committee

fTransfusion of platelets, frozen plasma, prothrombin complex concentrate, cryoprecipitate and/or fibrinogen concentrate; 17 patients received convalescent plasma and were not included in the count.

42

5. References

1. Schulman S, Kearon C. Definition of major bleeding in clinical investigations of antihemostatic

medicinal products in non-surgical patients. Journal of Thrombosis and Haemostasis. 2005;3(4):692-

694. doi:10.1111/j.1538-7836.2005.01204.x

2. Tang N, Bai H, Chen X, Gong J, Li D, Sun Z. Anticoagulant treatment is associated with decreased

mortality in severe coronavirus disease 2019 patients with coagulopathy. J Thromb Haemost.

18(5):1094-1099. doi:10.1111/jth.14817

3. Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality of adult inpatients with

COVID-19 in Wuhan, China: a retrospective cohort study. The Lancet. 2020;395(10229):1054-1062.

doi:10.1016/S0140-6736(20)30566-3

4. Sholzberg M, Tang GH, Negri E, et al. Coagulopathy of hospitalised COVID-19: A Pragmatic

Randomised Controlled Trial of Therapeutic Anticoagulation versus Standard Care as a Rapid

Response to the COVID-19 Pandemic (RAPID COVID COAG – RAPID Trial): A structured summary of a study protocol for a randomised controlled trial. Trials. 2021;22(1):202.

doi:10.1186/s13063-021-05076-0

5. Doi SA, Furuya-Kanamori L, Xu C, Lin L, Chivese T, Thalib L. Questionable utility of the relative risk in clinical research: a call for change to practice. J Clin Epidemiol. Published online November

7, 2020. doi:10.1016/j.jclinepi.2020.08.019

effect of therapeutic heparin vs prophylactic heparin on

Documents